It is well known that internal combustion engines have revolutionized transportation following their invention during the last decades of the 19th century. While others, including Benz and Gottleib Wilhelm Daimler, invented and developed engines using electric ignition of fuel such as gasoline, Rudolf C. K. Diesel invented and built the engine named for him which employs compression for auto-ignition of the fuel in order to utilize low-cost organic fuels. Development of improved diesel engines for use in transportation has proceeded hand-in-hand with improvements in diesel fuel compositions. Modern high performance diesel engines demand ever more advanced specification of fuel compositions, but cost remains an important consideration.
At the present time most fuels for transportation are derived from natural petroleum. Indeed, petroleum as yet is the world's main source of hydrocarbons used as fuel and petrochemical feedstock. While compositions of natural petroleum or crude oils are significantly varied, all crudes contain sulfur compounds and most contain nitrogen compounds which may also contain oxygen, but oxygen content of most crudes is low. Generally, sulfur concentration in crude is less than about 8 percent, with most crudes having sulfur concentrations in the range from about 0.5 to about 1.5 percent. Nitrogen concentration is usually less than 0.2 percent, but it may be as high as 1.6 percent.
Crude oil seldom is used in the form produced at the well, but is converted in oil refineries into a wide range of fuels and petrochemical feedstocks. Typically fuels for transportation are produced by processing and blending of distilled fractions from the crude to meet the particular end use specifications. Because most of the crudes available today in large quantity are high in sulfur, the distilled fractions must be desulfurized to yield products which meet performance specifications and/or environmental standards. Sulfur containing organic compounds in fuels continues to be a major source of environmental pollution. During combustion they are converted to sulfur oxides which, in turn, give rise to sulfur oxyacids and, also, contribute to particulate emissions.
Even in newer, high performance diesel engines combustion of conventional fuel produces smoke in the exhaust. Oxygenated compounds and compounds containing few or no carbon-to-carbon chemical bonds, such as methanol and dimethyl ether, are known to reduce smoke and engine exhaust emissions. However, most such compounds have high vapor pressure and/or are nearly insoluble in diesel fuel, and they have poor ignition quality, as indicated by their cetane numbers. Furthermore, other methods of improving diesel fuels by chemical hydrogenation to reduce their sulfur and aromatics contents, also cause a reduction in fuel lubricity. Diesel fuels of low lubricity may cause excessive wear of fuel injectors and other moving parts which come in contact with the fuel under high pressures.
Distilled fractions used for fuel or a blending component of fuel for use in compression ignition internal combustion engines (Diesel engines) are middle distillates that usually contain from about 1 to 3 percent by weight sulfur. In the past a typical specifications for Diesel fuel was a maximum of 0.5 percent by weight. By 1993 legislation in Europe and United States limited sulfur in Diesel fuel to 0.3 weight percent. By 1996 in Europe and United States, and 1997 in Japan, maximum sulfur in Diesel fuel was reduced to no more than 0.05 weight percent. This world-wide trend must be expected to continue to even lower levels for sulfur.
In one aspect, pending introduction of new emission regulations in California and other jurisdictions has prompted significant interest in catalytic exhaust treatment. Challenges of applying catalytic emission control for the diesel engine, particularly the heavy-duty diesel engine, are significantly different from the spark ignition internal combustion engine (gasoline engine) challenges due to two factors. First, the conventional three way catalyst (TWC) catalyst is ineffective in removing NOx emissions from diesel engines, and second, the need for particulate control is significantly higher than with the gasoline engine.
Several exhaust treatment technologies are emerging for control of Diesel engine emissions, and in all sectors the level of sulfur in the fuel affects efficiency of the technology. Sulfur is a catalyst poison that reduces catalytic activity. Furthermore, in the context of catalytic control of Diesel emissions, high fuel sulfur also creates a secondary problem of particulate emission, due to catalytic oxidation of sulfur and reaction with water to form a sulfate mist. This mist is collected as a portion of particulate emissions.
Compression ignition engine emissions differ from those of spark ignition engines due to the different method employed to initiate combustion. Compression ignition requires combustion of fuel droplets in a very lean air/fuel mixture. The combustion process leaves tiny particles of carbon behind which leads to significantly higher particulate emissions than are present in gasoline engines. Due to the lean operation the CO and gaseous hydrocarbon emissions are significantly lower than the gasoline engine. However, significant quantities of unburned hydrocarbon are adsorbed on the carbon particulate. These hydrocarbons are referred to as SOF (soluble organic fraction). A root cause of health concerns over diesel emissions can be traced to the inhalation of these very small carbon particles containing toxic hydrocarbons deep into the lungs.
While an increase in combustion temperature can reduce particulate, this leads to an increase in NOx emission by the well-known Zeldovitch mechanism. Thus, it becomes necessary to trade off particulate and NOx emissions to meet emissions legislation.
Available evidence strongly suggests that ultra-low sulfur fuel is a significant technology enabler for catalytic treatment of diesel exhaust to control emissions. Fuel sulfur levels of below 15 ppm, likely, are required to achieve particulate levels below 0.01 g/bhp-hr. Such levels would be very compatible with catalyst combinations for exhaust treatment now emerging, which have shown capability to achieve NOx emissions around 0.5 g/bhp-hr. Furthermore, NOx trap systems are extremely sensitive to fuel sulfur and available evidence suggests that they would need sulfur levels below 10 ppm to remain active.
In the face of ever-tightening sulfur specifications in transportation fuels, sulfur removal from petroleum feedstocks and products will become increasingly important in years to come. While legislation on sulfur in diesel fuel in Europe, Japan and the U.S. has recently lowered the specification to 0.05 percent by weight (max.), indications are that future specifications may go far below the current 0.05 percent by weight level.
Conventional hydrodesulfurization (HDS) catalysts can be used to remove a major portion of the sulfur from petroleum distillates for the blending of refinery transportation fuels, but they are not efficient for removing sulfur from compounds where the sulfur atom is sterically hindered as in multi-ring aromatic sulfur compounds. This is especially true where the sulfur heteroatom is doubly hindered (e.g., 4,6-dimethyidibenzothiophene). These hindered dibenzothiophenes predominate at low sulfur levels such as 50 to 100 ppm and would require severe process conditions to be desulfurized. Using conventional hydrodesulfurization catalysts at high temperatures would cause yield loss, faster catalyst coking, and product quality deterioration (e.g., color). Using high pressure requires a large capital outlay.
In order to meet stricter specifications in the future, such hindered sulfur compounds will also have to be removed from distillate feedstocks and products. There is a pressing need for economical removal of sulfur from distillates and other hydrocarbon products.
The art is replete with processes said to remove sulfur from distillate feedstocks and products. One known method involves the oxidation of petroleum fractions containing at least a major amount of material boiling above very high-boiling hydrocarbon materials (petroleum fractions containing at least a major amount of material boiling above about 550° F.) followed by treating the effluent containing the oxidized compounds at elevated temperatures to form hydrogen sulfide (500° F. to 1350° F.) and/or hydroprocessing to reduce the sulfur content of the hydrocarbon material. See, for example, U.S. Pat. No. 3,847,798 (Jin Sun Yoo, et al) and U.S. Pat. No. 5,288,390 (Vincent A. Durante). Such methods have proven to be of only limited utility since only a rather low degree of desulfurization is achieved. In addition, substantial loss of valuable products may result due to cracking and/or coke formation during the practice of these methods. Therefore, it would be advantageous to develop a process which gives an increased degree of desulfurization while decreasing cracking or coke formation.
Several different oxygenation methods for improving fuels have been described in the past. For example, U.S. Pat. No. 2,521,698 (G. H. Denison, Jr. et al.) describes a partial oxidation of hydrocarbon fuels as improving cetane number. This patent suggests that the fuel should have a relatively low aromatic ring content and a high paraffinic content. U.S. Pat. No. 2,912,313 (James B. Hinkamp et al.) states that an increase in cetane number is obtained by adding both a peroxide and a dihalo compound to middle distillate fuels. U.S. Pat. No. 2,472,152 (Adalbert Farkas et al.) describes a method for improving the cetane number of middle distillate fractions by the oxidation of saturated cyclic hydrocarbon or naphthenic hydrocarbons in such fractions to form naphthenic peroxides. This patent suggests that the oxidation may be accelerated in the presence of an oil-soluble metal salt as an initiator, but is preferably carried out in the presence of an inorganic base. However, the naphthenic peroxides formed are deleterious gum initiators. Consequently, gum inhibitors such as phenols, cresols and cresyic acids must be added to the oxidized material to reduce or prevent gum formation. These latter compounds are toxic and carcinogenic.
U.S. Pat. No. 4,494,961 (Chaya Venkat et al.) relates to improving the cetane number of raw, untreated, highly aromatic, middle distillate fractions having a low hydrogen content by contacting the fraction at a temperature of from 50° C. to 350° C. and under mild oxidizing conditions in the presence of a catalyst which is either (i) an alkaline earth metal permanganate, (ii) an oxide of a metal of Groups IB, IIB, IIIB, IVB, VB, VIB, VIIB or VIIIB of the periodic table, or a mixture of (i) and (ii). European Patent Application 0 252 606 A2 also relates to improving the cetane rating of a middle distillate fuel fraction which may be hydro-refined by contacting the fraction with oxygen or oxidant, in the presence of catalytic metals such as tin, antimony, lead, bismuth and transition metals of Groups IB, IIB, VB, VIB, VIIB and VIIIB of the periodic table, preferably as an oil-soluble metal salt. The application states that the catalyst selectively oxidizes benzylic carbon atoms in the fuel to ketones.
U.S. Pat. No. 4,723,963 (William F. Taylor) suggests that cetane number is improved by including at least 3 weight percent oxygenated aromatic compounds in middle distillate hydrocarbon fuel boiling in the range of 160° C. to 400° C. This patent states that the oxygenated alkylaromatics and/or oxygenated hydroaromatics are preferably oxygenated at the benzylic carbon proton.
U.S. Pat. No. 6,087,544 (Robert J. Wittenbrink et al.) relates to processing a distillate feedstream to produce distillate fuels having a level of sulfur below the distillate feedstream. Such fuels are produced by fractionating a distillate feedstream into a light fraction, which contains only from about 50 to 100 ppm of sulfur, and a heavy fraction. The light fraction is hydrotreated to remove substantially all of the sulfur therein. The desulfurized light fraction, is then blended with one half of the heavy fraction to product a low sulfur distillate fuel, for example 85 percent by weight of desulfurized light fraction and 15 percent by weight of untreated heavy fraction reduced the level of sulfur from 663 ppm to 310 ppm. However, to obtain this low sulfur level only about 85 percent of the distillate feedstream is recovered as a low sulfur distillate fuel product.
U.S. patent application Publication Ser. No. 2002/0,035,306 A1 (Gore et al.) discloses a multi-step process for desulfurizing liquid petroleum fuels that also removes nitrogen-containing compounds and aromatics. The process steps are thiophene extraction; thiophene oxidation; thiophene-oxide and dioxide extraction; raffinate solvent recovery and polishing; extract solvent recovery; and recycle solvent purification.
The Gore et al. process seeks to remove 5–65% of the thiophenic material and nitrogen-containing compounds and parts of the aromatics in the feedstream prior to the oxidation step. While the presence of aromatics in diesel fuel tends to suppress cetane, the Gore et al. process requires an end use for the extracted aromatics. Further, the presence of an effective amount of aromatics serves to increase the fuel density (Btu/gal) and enhance the cold flow properties of diesel fuel. Therefore it is not prudent to extract an inordinate amount of the aromatics.
With respect to the oxidation step, the oxidant is prepared in situ or is previously formed. Operating conditions include a molar ratio of H2O2 to S between about 1:1 and 2.2:1; acetic acid content between about 5 and 45% of feed, solvent content between 10 and 25% of feed, and a catalyst volume of less than about 5,000 ppm sulfuric acid, preferably less than 1,000 ppm. Gore et al. also discloses the use of an acid catalyst in the oxidation step, preferably sulfuric acid. The use of sulfuric acid as an oxidizing acid is problematic in that corrosion is a concern when water is present and hydrocarbons can be sulfonated when a little water is present.
According to Gore et al. the purpose of the thiophene-oxide and dioxide extraction step is to remove more than 90% of the various substituted benzo and dibenzo thiophene-oxides and N-oxide compounds plus a fraction of the aromatics with an extracting solvent that is aqueous acetic acid with one or more co-solvents.
U.S. Pat. No. 6,368,495 B1 (Kocal et al.) also discloses a multi-step process for the removal of thiophenes and thiophene derivatives from petroleum fractions. This subject process involves the steps of contacting a hydrocarbon feed stream with an oxidizing agent followed by the contact of the oxidizing step effluent with a solid decomposition catalyst to decompose the oxidized sulfur-containing compounds thereby yielding a heated liquid stream and a volatile sulfur compound. The subject patent discloses the use of oxidizing agents such as alkyl hydroperoxides, peroxides, percarboxylic acids, and oxygen.
WO 02/18518 A1 (Rappas et al) discloses a two-stage desulfurization process which is utilized downstream of a hydrotreater. The process involves an aqueous formic acid based, hydrogen peroxide biphasic oxidation of a distillate to convert thiophenic sulfur to corresponding sulfones. During the oxidation process, some sulfones are extracted into the oxidizing solution. These sulfones are removed from the hydrocarbon phase by a subsequent phase separation step. The hydrocarbon phase containing remaining sulfones is then subjected to a liquid-liquid extraction or solid adsorption step.
The use of formic acid in the oxidation step is not advisable. Formic acid is relatively more expensive than acetic acid. Further, formic acid is considered a “reducing” solvent and can hydride certain metals thereby weakening them. Therefore, exotic alloys are required to handle formic acid. These expensive alloys would have to be used in the solvent recovery section and storage vessels. The use of formic acid also necessitates the use of high temperatures for the separation of the hydrocarbon phase from the aqueous oxidant phase in order to prevent the appearance of a third precipitated solid phase. It is believed this undesirable phase can be formed due to the poor lipophilicity of formic acid. Therefore at lower temperatures, formic acid cannot maintain in solution some of the extracted sulfones.
U.S. Pat. No. 6,171,478 B1 (Cabrera et al.) discloses yet another complex multi-step desulfurization process. Specifically, the process involves a hydrodesulfurization step, an oxidizing step, a decomposition step, and a separation step wherein a portion of the sulfur-oxidated compounds are separated from the effluent stream of the decomposition step. The aqueous oxidizing solution used in the oxidizing step preferably contains acetic acid and hydrogen peroxide. Any residual hydrogen peroxide in the oxidizing step effluent is decomposed by contacting the effluent with a decomposition catalyst.
The separation step is carried out with a selective solvent to extract the sulfur-oxidated compounds. Per the teachings of Cabrera et al. the preferred selective solvents are acetonitrile, dimethyl formamide, and sulfolane.
A number of solvents have been proposed for removing the oxidized sulfur compounds. For example, in U.S. Pat. No. 6,160,193 (Gore) teaches the use of a wide variety of solvents suitable for use in the extraction of sulfones. The preferred solvent is Dimethylsulfoxide (DMSO).
A study of a similar list of solvents used in the extraction of sulfur compounds was published by Otsuki, S.; Nonaka, T.; Takashima, N.; Qian, W.; Ishihara, A.; Imai, T.; Kabe, T. “Oxidative Desulfurization of Light Gas Oil and Vacuum Gas Oil by Oxidation and Solvent Extraction” Energy & Fuels 2000, 14, 1232. That list is displayed below:                N,N-Dimethylformamide (DMF)        Methanol        Acetonitrile        Sulfolane        
Gore states that there is a relationship between the solvent's polarity with the solvent's extraction efficiency. All of the solvents listed in the patent and the paper are desirably immiscible with the diesel. They are all characterized as either polar protic or aprotic solvents.
There are several deleterious effects arising from the use of the above-mentioned solvents. While DMSO and sulfolane are good solvents for extractions, there is a tremendous risk that any traces of these solvents left in the product could dramatically increase the sulfur concentration in the diesel product. For example, even residual DMSO in the final product at a concentration of 37 ppmw would impart a sulfur concentration the final diesel fuel of 15 ppmw. Similar detrimental effects can result from the use of acetonitrile, triethanolamine, and DMF which contain nitrogen atoms. Trace levels of these solvents would dramatically increase the nitrogen concentration in the final product.
The above solvents are not particularly selective for sulfur, as they will also remove aromatics, particularly monoaromatics since these species are likely to be the most polar components of a diesel fuel. On the surface, it would appear to be beneficial to enrich the diesel fuel with saturates (paraffins) by removing these aromatics, which achieves a higher cetane number in the fuel. The downside is that the extracting solvent's stream size would swell dramatically and would contain these monoaromatics of some value which must be recovered. For example, it is known from the above cited article that DMF can extract unoxidized dibenzothiophenes, but it also removes a substantial portion of the oil. A significant effort would be needed to recover the hydrocarbon and not co-recover the dibenzothiophenes.
Another concern with the use of the above solvents is the boiling point. A higher boiling point would make it difficult to separate traces of the solvent from the final product by a flash. A flash here would result in taking some of the lower boiling diesel components with it. For example, DMSO has a boiling point of 189° C. or 372° F., and DMF has a boiling point of 153° C. or 307° F. The initial boiling point of a diesel fuel is typically below the boiling points of these two solvents.
Toxicity is another issue. While DMSO is technically a low toxicity solvent, it is classified as a “super-solvent” which can dissolve a wide variety of compounds. Skin contact of this DMSO solution will rapidly cause the solute to be adsorbed through the skin, which is one of the characteristics of DMSO. DMF is a liver toxin and a suspected carcinogen. Acetonitrile is also quite toxic.
DMF is not thermally stable enough to be distilled under atmospheric pressure. At the ambient pressure boiling point of DMF, degradation also occurs to give carbon monoxide and dimethyl amine (Perrin, D. D.; Armarego, W. L F. Purification of Laboratory Chemicals, 3rd Edition, Pergamon Press, Oxford, 1988, page 157). Vacuum distillation is therefore required.
Gore in the '193 patent points out the shortcomings associated with methanol, i.e. the methanol has about the same density as the typical hydrocarbon fuel. Based on a process of elimination, methanol appears to be a good solvent in terms of its boiling properties, and the fact that it will not leave behind nitrogen or sulfur. However, a significant fraction of the total hydrocarbon will also be extracted into the methanol layer. Methanol is also disadvantaged by the fact that it does not rapidly separate from the diesel.
In view of the above, it is clear that there is a need for a less complex, economic distillate or diesel desulfurization process that does not employ the use of toxic solvents such as acetonitrite or DMF, super-solvents such as DMSO or hard to separate solvents such as DMF.
The present invention provides for a relatively simple process wherein a portion of the oxidized sulfur containing and/or nitrogen-containing organic compounds contained in a hydrocarbon feedstock are extracted simultaneously during an oxidation process step and subsequently separated via a decantation or phase separation step. This phase separation results in less sulfur and nitrogen species to be removed further downstream via an extraction step. Further, the process of the present invention provides for the use of a single solvent, acetic acid for both the oxidation step, and an extraction step; thereby permitting the use of only one regenerator tower to regenerate the acetic acid for both the oxidation step and the extraction step. In a specific embodiment of the present invention, the invention provides for the use of a reduced amount of expensive oxidizing agent in the oxidation step.